Researches And Analysis On Email Virus Propagation Model Based On Complex Network

Currently, computer virus poses a serious threat to Internet. The high level of destructivity has aroused wide concern by corporations and individuals. As we know, computer viruses are programes which probably damages the computers by erasing data, stealing information, or modifying the normal data, even making whole network system collapse. In recent years, one new type virus named email virus spreads in the Internet. Once the virus is actived successfully by user, it will easily send out its virus copies to other email users in the list of the users' address book. Due to the facility, the creaters of virus mostly tend to the email measures to spread email virus. Thus, researchs on email virus propagation characteristics, spreading model and controlling schemes have become urgent problems in network information security field.Pioneering achievements on epidemic spreading are mostly based on epidemiological theory without considering topological aspects of the network. In fact, computer virus propagation greatly depends on topological structures of networks. Recently, the rapid development of complex networks exhibits many new sights, means and appoarches for studying virus epidemic model. In this paper, the intensive research is developed on email virus speading model based on complex network theory. The fruitful simulations demonstrate that proposed virus spreading models are reliable and effective.First, after make anlysis for email virus spreading mechanism, spreading process, current research in the worldwide and development trend, this paper gaved a recapitulative introduction for the basal knowledge about complex network from ogrinal, development, related notion and recent achievements. Then, some classical virus spreading models are introduced and anlysed.Second, the paper attempts to develop an acurate mathematical model to reflect the essential situation, which is based on homogeneous network. Our contributions are: The models are two phase virus models according to the time of appearing anti-virus technique. More importantly, the paper incorporates two new parameters: User Vigilance and Removing Time. It reveals human's positive behavior plays an important role in contolling email virus diffusion. Simulations prove that, if system with high user vigilance and small removing time, virus spreading can be easily defeated in spite of infected computers accumulated a high level. Furthermore, the system may only suffer a small damage.Third, the propagation of email virus usually depends on the behavior of sending and receiving email in the Internet. Recent researches show that, complex system (such as the physical Internet, the WWW) displays a highly heterogeneous nature and large connectivity and performance fluctuations. The function of degree distribution complies with power law. Scientists call this topological structure as scale-free network. Growth and preferential attachment are its typical characteristics. In view of these facts, it is necessary that one email virus model accounts for the complexity of the network at various factors: degree distribution, degree of nodes, average degree of network and users' coefficient. In this paper, the email virus spreading model based on the scale-free network was proposed. By mathematical calculation and analysis, the density of infected users is obtained, which is related with the degree of nodes. Experimental results indicate that the expressions (the density of infected users) are appropriate and correct.